STA 4702 Multivariate Statistical Methods (3)

Spring 2003


Notes:

  1. Final course notes posted.
  2. The MANOVA Exercise is not required but provided for those who wish to practice.
  3. The Final Quiz Prototype is posted below.
  4. Data and SAS program for Manova exercise are in DATASETS and SAS Programs.
  5. Check SAS programs for in class examples.
HW # 1-Friday 1/31 HW # 2.-Due Friday 2/14 HW # 3 Due 3/17
HW # 4 Monday 4/21. Final Quiz Prototype. MANOVA Exercise

 

DATASETS SAS Programs R Programs


 

Syllabus:

Below is the list of topics in the order to be presented. Please recognize that this is a very ambitious topics list and will require the student to keep up with readings and homework assignments.

WEEK LECTURE NOTES

ASSOCIATED MATERIAL

TOPICS
1

Concepts

Computing

SAS Intro

R Intro

Basic Concepts and Matrix Algebra
2 Concepts   Univariate normal, t, chi-squared, F and multivariate normal distributions.
3 PCA Notes PCA SAS Principal Components Analysis
4 CanCorr Notes CanCorSAS Canonical Correlations Analysis
5 Factor Notes Factor SAS Factor Analysis
6 Discriminate 1   Discriminant Analysis/Pattern Recognition
7 Discriminate 2 Missing Obs1 Non-Parametric Discrimination and Neural Networks
8 Correspondence Analysis   Correspondence Analysis (3/17)
9 Path Analysis   Path Analysis and Structural Equations (3/19)
10 Cluster Analysis   Cluster Analysis (3/21-3/28)
11 Matrix Regression   Multivariate Regression (4/2)
12 Multivariate Hypotheses   Multivariate Analysis of Variance (4/4-4/7)
13 Repeated Measures   Analysis of Repeated Measures (4/9-4/11)
14 RM for Treatment Designs   Repeated Measures and Mixed Models (4/14-4/18)
15 Mixed Models   Additional Topics and catch-up (4/21-4/23)

  1. Instructor: Kenneth M. Portier PhD., Associate Professor of Statistics, (portier@ufl.edu)
  2. Office: 522 McCarty Hall C
  3. Telephone: 392-3067 (with answering machine)
  4. Lectures: M-W-F 2nd Period (8:30am-9:30am)
  5. Office Hours: 10-12am Wednesday and 2-4pm Friday or by appointment.
  6. Prerequisites: STA 3024 or STA 4210 or STA 4322 or STA 6127 or STA 6167 or consent of instructor.

Return to Top Return to the IFAS Statistice Home Page

Catalog Description:

Review of matrix theory, univariate normal, t, chi-squared, F and multivariate normal distributions. Inference about multivariate means including Hotelling's T squared, multivariate analysis of variance, multivariate regression and multivariate repeated measures. Inference about covariance structure including principal components, factor analysis and cannonical correlation. Multivariate classification techniques including discriminant and cluster analyses. Additional topics at the discretion of the instructor, time permitting.

Course Objective:

  1. To introduce statistical inference on means and variances in a multivariate normal setting.
  2. To introduce exploratory multivariate analysis techniques.

Course Texts:

Grading:

Since this is an applied methods course, the emphasis will be on applications of the methods discussed in class. Graded homework will be assigned each week. Homeworks will require the use of the computer software. The SAS System software will be the primary computing tool, but some examples in S+ (or R ) may be used in those situations that SAS doesn't handle well. We will discuss how student actually get access to SAS, S+ or R in the first week of classes. Some student may wish to acquire their own license for SAS to do the course homeworks. R is publically available and S+ is available in the Statistics Department and other departments around campus.

Short reports on the conclusions and/or findings of the analysis will be part of the homework requirements. You cannot pass this class without doing the homeworks.

One or more quizzes may be given but they will not constitute the bulk of course credit.

Course Notes:

Lectures will be accompanied with instructor-prepared powerpoint slides. Powerpoint slides will be converted to pdf format and posted on this web site. The pdf format will contain 4 slides per page providing a consise record of what was discussed in lecture. Lecture slides should be available prior to the lecture but is not quaranteed. Adobe Acrobat Reader is required to view these files. If you do not have Acrobat, it may be obtained from the Adobe web site.